1. Field of the Invention
The invention relates in general to a photolithography process, and more particularly, to a method of figuring the exposure energy.
2. Description of the Related Art
For the enhancement of the integration for integrated circuit, the device dimension of the integrated circuit has to be reduced. Photolithography process is one of the most crucial semiconductor fabrication processes. The properties related to the metal-oxide-semiconductor (MOS) device structure such as the pattern of each thin film and the area doped with dopants are determined according to the photolithography process.
In the conventional critical dimension (CD) control for photolithography process, the exposure energy is compensated according to the measured results of critical dimension after performing the exposure process. Therefore, the wafers in the next batch can have a critical dimension closer to a target value.
However, the conventional method performs compensation on the exposure energy of the wafers after obtaining the CD measurement results of the exposure pattern. Therefore, the current batch of wafers has to be sacrificed. In addition, the exposure condition for a new product can only be referred from that of the same products or similar products. The exposure energy is given based on experience. The photomask CD deviation and on-line product CD baseline are not considered in this method it is often that the CD of the final exposed pattern cannot meet the target value, or even exceeds the deviation tolerance. Thus, the exposure condition of the product has to be modified. Moreover, the conventional method cannot predict how the material (such as photoresist material) used in different batch affects the CD, so that an advanced compensation cannot be made.
The invention provides a method of figuring an exposure energy that predicts the compensation in advance to reduce the deviation of CD of the final exposure pattern. The CD of the exposure pattern thus further approaches the target value.
The method of figuring an exposure energy resolves the problem of unable to figure the exposure energy for new product in the prior art.
In the method of figuring exposure energy provided by the invention, a required exposure energy is calculated according to a photomask critical dimension of a layer to be exposed (also called as xe2x80x9can exposing layerxe2x80x9d hereinafter). From the thickness of a layer formed before the exposing layer, a first critical dimension (CD) deviation is obtained. A first energy compensation is obtained by calculating the first CD deviation. Whether the deviation of photoresist sensitivity between two sequential batches is less than 1% is checked. If the deviation of photoresist sensitivity between two sequential batches is less than 1%. an exposure energy applied to the exposing layer is obtained by adding the required exposure energy with the first energy compensation. Otherwise, a second CD deviation is obtained according to the deviation of photoresist sensitivity of two sequential batches, and a second energy compensation is commutated from the second CD deviation. The sum of the required exposure energy, the first energy and the third energy is the exposure energy applied to the exposing layer.
The invention compensates the energy in advance to reduce the CD deviation of the exposure pattern, so that the CD of the resultant exposure pattern is closer to the target value.
In the invention, as the exposure energy is compensated in advance, so that sacrifice of wafers for the current batch because only wafers of the next batch can be compensated in the conventional method is avoided.
Therefore, the invention accurately figures the exposure energy of a new product, so that the problem for not able to predict the exposure energy in the prior art is resolved.
Both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.